opovid 0.1.7

Core runtime for the Opovid testing library.

Opovid Python Runtime

This directory contains the Python runtime for Opovid.

The first Python slice ships the framework-agnostic core plus the shared remembered-state ability:

• Actor and the preferred User facade
• task, question, and check helpers
• with_retry for wrapping flaky or eventually-consistent tasks
• nested execution reports and timeline events
• render_timeline, timeline_to_json, and summarize_failure diagnostics
• listener hooks for step boundaries
• RememberTheState for small actor-local state

Install Locally

From this directory:

python3 -m pip install -e .

Basic Usage

from opovid import Actor, RememberTheState, remembered_by, question, task

casey = Actor.named("Casey").can(RememberTheState.using_defaults())

casey.attempts_to(
    task("remembers a subscription", lambda actor: remembered_by(actor).remember("plan", "pro"))
)

plan = casey.asks_for(question("reads the selected plan", lambda actor: remembered_by(actor).recall("plan")))

assert plan == "pro"

Module Layout

The Python runtime follows the same concept boundaries as the TypeScript, Java, and Ruby runtimes:

• actor.py, user.py, and attempt_chain.py own the user/actor flow.
• task.py, question.py, and check.py keep the executable contracts separate.
• execution.py and execution_options.py own reports, timelines, errors, listeners, and execution options.
• timeline.py owns text/JSON timeline rendering and failure summaries.
• retry.py owns retry task wrappers.
• actor_memory.py, remember_the_state.py, and memory.py provide the shared remembered-state ability.
• core.py and __init__.py are public re-export entrypoints.

The Python runtime does not currently include the scenario spec module available in the TypeScript, Java, and Ruby runtimes.

Run Tests

From this directory:

PYTHONPATH=src python3 -m unittest discover -s tests

Lint And Format

Ruff is the Python runtime's linter and formatter.

python3 -m pip install -e ".[dev]"
ruff check .
ruff format --check .

Publish

The Python runtime publishes as opovid on PyPI.

Tagged releases are handled by the root GitLab release pipeline. The Python release job builds the source distribution and wheel, checks them with Twine, and uploads with PyPI Trusted Publishing through GitLab OIDC. Configure the PyPI trusted publisher for the sunrise.water/opovid GitLab project, the top-level .gitlab-ci.yml workflow, and the release environment before the first publish.

Browser flows remain intentionally centered in the TypeScript runtime with Playwright. The Python runtime is currently focused on service, request, job, and plain unit/integration flows.